Necron

Last night I installed the plugin and to my surprise, everything was fine. Drives went to sleep properly, VM ran fine, etc. When I powered on my main PC this morning however, shit hit the fan again. I tried to resume the VM from pause state (how I usually handle it) and I noticed errors popping up from the log: Then, after the end of that snippet was a bunch of bunker: error: no export of file: yadda yadda yadda's. After a hard-off and cooldown I rebooted it and verified that the files in the error reports were in-tact. It seems like there was a big hissy-fit between the parity drive and the SAS HDD (disk5) which started upon my resuming the SAS SSD (VM). I'm removing the spindown plugin for now and I'll see if I can go a week without issues. I'm cool to try any tests you want I read the description of the SAS-Sata adapter and it mentions "Before buying, please check the manual to see if it supports SATA mode because it is not compatible with all SAS hard drives." This could all be the adapter's fault. I'll have a think and see what would be the cheapest/easiest way to get the adapter out of the setup. I may have to get an SFF 8088 to 4xSFF 8482 cable just to connect the two SAS drives.

Well, after the parity completed I cautiously booted up the VM with unraid log window on my other screen. No problems! I continued to use the VM, start/stop/pause/hibernate - no problems. I'll install the latest version of spindown, restart the box and try the same things again. Yep! Image attached. Ebay link here If I run my tests after reinstalling the plugin and the problems surface again I'd be surprised if it's a cable problem, but it never hurts to check. I'll go and have a gander when I need to hard off the machine.

A) It's currently doing a parity check as I was forced to unsafe shutdown. Currently SAS Spindown is uninstalled, but I'll reinstall it after parity finishes. B) Attached HP Microserver Gen8 // Xeon E3-1200 v2 // 16GB RAM ----- Four Sata drives inside the Gen8 enclosure [2:0:0:0] disk ATA ST2000DM001-1E61 CC45 /dev/sdb 2.00TB [3:0:0:0] disk ATA ST2000DM001-1E61 CC45 /dev/sdc 2.00TB [4:0:0:0] disk ATA ST2000DL003-9VT1 CC3C /dev/sde 2.00TB [5:0:0:0] disk ATA WDC WD10EACS-00Z 1B01 /dev/sdf 1.00TB ----- LSI SAS2008 in PCI Express - Eight external drives [1:0:0:0] disk HITACHI HUS72403CLAR3000 C370 /dev/sdd 3.00TB -- SAS to Sata [1:0:1:0] disk ATA WDC WD30EZRX-00D 0A80 /dev/sdg 3.00TB [1:0:2:0] disk HITACHI HUSRL402 CLAR200 C140 /dev/sdh 200GB -- SSD SAS to Sata [1:0:3:0] disk ATA Hitachi HDS5C302 A580 /dev/sdi 2.00TB [1:0:4:0] disk ATA Hitachi HDS5C302 A580 /dev/sdj 2.00TB [1:0:5:0] disk ATA WDC WD30EZRX-00D 0A80 /dev/sdk 3.00TB -- Parity [1:0:6:0] disk ATA Hitachi HDS5C302 A5C0 /dev/sdl 2.00TB [1:0:7:0] disk ATA Hitachi HDS5C302 A580 /dev/sdm 2.00TB The 200GB SSD (and previously the 120GB Intel SSD) are passed through to the Windows 10 VM (/dev/disk/by-id/scsi-35000cca0132a3e80). edit: Also all the CPU cores and 10752MB RAM, network bridge BR0 tower-diagnostics-20210828-0559.zip

G'day, I've been having a specific problem with my box for about two months now and essentially it's started after putting some SAS drives in. I don't want to point fingers yet, I'd just like some thoughts on it Server runs fine, things spin up and down ok until I start the windows 10 VM on my SSD. I have two log snippets here during the time things went to hell, first is with the SSD being an intel 120G sata and second being the hitachi 200G SAS (with SAS-Sata adapter) that replaced the intel SSD thinking it may have been on the way out. The other drives in the box are Sata (including parity) except for two hitachi 2TB SAS (with SAS-Sata adapters). Log 1: Intel 120G SSD This log has a few drives spitting read errors, but the consistent problem between all the box crashes is disk0, the parity. Log 2: Hitachi 200G SSD General behavior at this point after each log is infinitely expanding read errors on disk0, CPU being nearly pinned, system doesn't respond to array stop or shutdown and has to be hard off'd. As mentioned before, system is generally stable without using windows 10 VM. Said VM is passthrough'd to the SSD, if passing through to a drive with SAS drives in the mix is considered a "no-no" that would be nice to know. Anything else I'll be happy to follow up with if requested.